THE MESOLITHIG INDUSTRIES OF MOTHER SILLER'S CHANNEL ... · 1. The Mother Siller's site appears to have been a camping site during the meso-lithic period, possibly afte B.Cr whe 500

THE MESOLITHIG INDUSTRIES OF MOTHER SILLER'S CHANNEL, CHRISTCHURCH,

AND THE NEIGHBOURING AREAS

EXCAVATION REPORT By SUSANN PALMER

WITH APPENDIX ON ANALYSIS OF SOIL SAMPLES By I. W. CORNWALL (Inst, of Archaeology)

I. EXCAVATION OF THE MOTHER SILLER'S CHANNEL SITE The Site

The site is by the Stanpit Nature Reserve on Christchurch Harbour (N.G.R. SZ 16859181). The excavated area is on the lowest part of a sandy prominence known as Crouch Hill. The area is turf-covered with small clumps of low bushes. Immediately round the site are low-lying marshes, but across the harbour is the headland of Hen-gistbury Head with Warren Hill as the highest part (figs. 3 and 4). Surface collecting

Fig- 3

9

PROCEEDINGS FOR THE YEAR I97O

G.N

M.N.

' f I ! 1 1

C.G. H E N . G I S T B U R Y H E A D

M O T H E R S I L L E R ' S C H A N N E L

Fig. 4. Mother Siller's Channel, plan of excavations.

IO

THE MESOLITHIC INDUSTRIES

of mesolithic artifacts in the area by Mr. and Mrs. A. Cotton first drew attention to the site. It was decided to undertake excavations in order to establish the nature of the site.

The site is 4.3 ft. above Ordnance Datum Newlyn. Tides in the harbour are, to some extent, affected by the sand-spit across the harbour mouth. As H.W.O.S.T. is 1.6 ft. above O.D., the site is only flooded by exceptionally high tides or in storms. No evidence of gleying was found in the trenches excavated and it can therefore be assumed that the site was anciently outside the reach of tidal water, although by c.

2 F T

. 6 M

S A N D L A Y E R 2

A R T I F A C T C L U S T E R X / •

/

A N I M A L , ->-H O L E S ' ' - — y~-

<?

I /

• • • • t * * • •

H E A R T H . , ,. S T O N E S . ^ A J ' B U R N T

SA ND

h - J Y E L L O W J J J j J , C L A Y

x - J - I J . . -

» * • »

* X • • # . • • : • <

X

O

Fig. 5. Mother Siller's Channel, plan of Trench i .

I I


4000 B.C. the sea was generally at the present level (Godwin and Willis 1959, 68). The natural stratigraphy of the area is indicated in the section taken immediately

by the edge of the shore (fig. 7). The horizons are those of a humic podsol developing in sand, but apparently lacking the iron concretions in the illuvial horizon, unless they are below the depth excavated. The boundaries of the layers are clearly defined where undisturbed. The topsoil is a dark brown colour and is matted by grass roots. Darker bands of humic sand occur within this zone. Below this is a pale grey eluvial horizon of sand, followed by a dark grey layer and then a light brown sand layer. At the base of the sand is a brown, fine, sandy gravel such as occurs just beyond tide marks. During surface collecting, artifacts have been found protruding from the base of layer 2 and from the dark grey sand. The dark grey sand is possibly the (A) horizon of the ancient surface with the light brown sand of layer 4 as a leached horizon. These observations made in the field are substantiated by the laboratory examination of the samples taken from each of the layers (Appx.).

THE EXCAVATIONS Trench 1

The stratigraphy of this trench is indicated in figure 8. It corresponds to the section along the shore edge, discussed above. Artifacts occurred in all layers but the majority were found towards the base of layer 2 (the light grey sand). Several runs of burrowing animals were encountered.

Two hearths with large areas of burnt, concreted sand round them were found at the base of layer 2. The one hearth consisted of two slabs of Purbeck limestone placed together and the other consisted of one big slab of limestone with an adjoining area of compact yellow clay (fig. 5). The sand surrounding the hearths was stained dark grey. Only a few very small pieces of charcoal could be found, and it is surmised that the wood was burnt to fine ash which subsequently became absorbed by the sand.

Artifacts were spread over the whole area of the trench, except for a number of small isolated clusters of waste flakes. Two microliths came from the sand round the hearths. The flint artifacts from Trench 1 were all of a mesolithic facies. One or two pieces of modern glass came from the animal burrows.

Trench 2 Three large hearth-stones, also of Purbeck limestone, were found in this trench.

Here again, the sand round the stones was burnt to a great depth and was concreted as a result of the fusion of the silica (fig. 6).

Several post-mesolithic features were encountered in Trench 2. The section drawings (figs. 9 and 10) show two animal burrows, one of which contained a piece of medieval pottery. The light grey sand of layer 2 is replaced by very dark brown sand, reaching right down to the top of the hearth-stones. Without excavating a larger area,

12

H


2 FT D .6M

O

S A N D

Fig. 6. Mother Siller's Channel, plan of Trench a.

13


'll I l l | |IM| 11III III || l ] | III|TT|K

: . . . ' . ' . * • * . * : t < • : ~ ~

1 11 I I I I n 1 1 1 7 1 1 1 1 1 1T I i , I I I I I M I I 1 1 1 I I I 1 l

1 11 1 1 I I I I I I I I \ 3 M M

' l i I 1 I I l I I I I I I I I I I M

4 _ _ _

O 0 ° * o o a o . « 0 o w

0 ° t °: o° .* 5 o 0 * 0

0 1 FT 1 I .7 0 .3 M

Fig. 7. Mother Siller's Channel. Natural stratification

Fig. 8. Section on south-east face of Trench 1.

1. Dark brown sand with humic black bands.

2. Light grey leached sand.

3. Dark grey sand.

4. Light brown sand.

5. Fine gravel in brown sand matrix.

6. Concreted burnt sand.

7. Very dark brown sand — recent intrusive feature.

8. Slabs of Purbeck limestone, cracked by heat.

9. Compact yellow clay.

10. Recent animal burrows.

P i . Sherd of medieval pottery.

P2. Sherds of Saxon pottery.

14


, . i i U . v . M i r - - - - ' < - - / - . -C8 l . i . 0,^.s-V,'fi", fr, (.-. f,i f.-'<S N i l . ' I l l l ' " " ^ ! f i , 6 ,••••

- • • • V " 5 • . 0 * 0 <» • •

2 F T

. 6 M

fllHmHMf'irTTTTTTTrTTTlH

•I . • • • ' • • • • »

Vl I I I I • • N l I •. • • " J I M I l L X I J J L L I J L I

• 6 I i ' • • ' ' ' ' i i i I I M I I J ' / V

, [ I I I I I I I I I I I 11 I I I I I i——I I -t I I I I I I I I •

2FT

.6 M Figs. 9 - 10. Mother Siller's Channel. Section faces, Trench 2.

r5


it was not possible to determine the exact nature of this intrusive feature, but it appears as though at some time a hollow had been scooped out of the sand embankment from the direction of the foreshore. Several small sherds of Saxon pottery and one very small sherd of possible Bronze Age date were found in the dark sand. The layers below the hearths corresponded with the layers along the foreshore. Very few artifacts were found in Trench 2 and of these only a few pieces are of interest. Apart from the pottery, nothing was found which could be ascribed to any period other than the mesolithic.

Finds

T R E N C H I

• LAYER I LAYER 2 LAYER 3 T O T A L

Microliths: Obliquely blunted (Ai) 1

Blunted along one side (Form. Bi) 1

One side plus base (C) 1 1

Scrapers: Side I 1

Convex 1

Concave 1

Blunted flakes 2 2

Gravers 1

Saws 1

Blade segments, plain i 2

Blade segments, retouched 1

Retouched flakes i 2

Retouched blades 1

Utilised flakes 3 4 7 Utilised blades 1

Abraded pebbles 1

Micro-blades, notched 2 2

Cores: i platform, part perimeter retouched 2 2 2 6 2 platforms, parallel I 1

2 platforms, intermediate angle 2 2

3 platforms I 1

4 platforms I 1

2 platforms, intersecting 2 1 3 unclassified, multi-platform 3 1 4

Core-trimming flakes 8 J 9 3 30 Waste flakes 89 109 198 Waste blades 5 5

Totals " 5 '52 11 278

16


T R E N C H 2

TOTAL

Scrapers: Convex Gravers Retouched flakes Utilised flakes Abraded pebbles Cores: 2 platform, right angle

Discoidal Unclassified, multi-platform

Core-trimming flakes Axe trimming flake Waste flakes Waste blades

1 1 2

4 2 1 1 1

9 1

65 2

90

Trenches 1 and 2 - Total artifacts: 368

Surface Collecting on the Foreshore One tranchet axe trimming flake and two microliths from the shore immediately by

the excavation are of significance and will be included in the totals of artifacts from the site. The microliths are: one scalene triangle (Form Dib) and one of Form B2 with two sides blunted. This makes a total of five microliths from the site.

Portland Chert Artifacts Four small artifacts made of Portland Chert were found in Trench 1. They are:

two waste flakes; one micro-blade and one oblique trimming flake from a tiny micro-core.

Artifact Analysis Although the assemblage from Mother Siller's Channel is small, it can be regarded

as representative of the lithic output of groups of coastal wanderers in the area. Microliths The ratio of microliths to macrolithic tools (retouched or utilised) is

1:6.6. This is more than the ratio of scrapers to macroliths (1:8.2). No deductions can be made about the typology of the microliths as so few have been found; it should be noted that the small assemblage includes geometric as well as non-geometric forms.

Macrolithic Assemblage The only other tools in the small assemblage are four scrapers, two gravers, one saw, two blunted flakes and six retouched blades of flakes. If we count the utilised flakes and blades and the abraded pebble also as tools, the ratio of tools to waste is 1 :io.4, which is a ratio one could expect on a temporary camping site where local beach material was utilised for immediate artifact requirements.

Axes Although no axes were found, the presence of two tranchet axe trimming flakes indicate that they were made on the site. Judging from the dimensions of the trimming

B 17


flakes, one axe would have been very small, only suitable for cutting down marsh vegetation. The other axe would have been sturdier.

Axes and picks appear to be a regular feature in the coastal industries of Hampshire and the adjoining counties.

SUMMARY

1. The Mother Siller's site appears to have been a camping site during the meso-lithic period, possibly after 5000 B.C. when the sea may have been a short distance from the site. Earlier sites on the ancient shoreline will now be submerged.

2. The stone industry is small but the ratio of the various artifacts to each other is well balanced. The amount of waste far exceeds that of tools. This is consistent with the economy of wandering folk, staying for a short while in a favourable spot with an adequate supply of raw material.

3. The slabs of Pur beck limestone for the hearths and the Portland chert artifacts are of particular interest as they indicate folk-movement from Dorset, possibly from the Isle of Purbeck, in the case of the hearth-stones, and from the Isle of Portland or the Fleet. The nearest source of Purbeck limestone to Mother Siller's Channel is a walking distance of approximately 15 miles away (fig. 3). It is difficult to explain why fairly heavy stone slabs were carried this distance, except perhaps for the fact that the stones from the Eocene Beds of the Christchurch area are inclined to crack apart rather quickly when subjected to heat.

4. Tranchet axes were made and used by the mesolithic coastal dwellers. 5. The microliths include geometrical forms.

I I . T H E MESOLITHIC OF THE CHRISTCHURCH AND BOURNEMOUTH AREA

Hengistbury Head, Bournemouth Mother Siller's Channel and the Hengistbury Head neighbourhood were probably

part of the same settlement pattern during the early Flandrian period prior to the complete inundation of the present harbour. During the later phases the two sites would still have been within easy reach of each other.

Mesolithic artifacts have been found by Draper (Rankine 1956, 37) in a fairly concentrated area of Warren Hill, along the footpath just north of the Late Palaeolithic site excavated by Mace (1959) and recently by Campbell (unpublished).

Since then a careful and systematic examination of the whole of Hengistbury Head and surrounding fields has been carried out by the Marchese Alessandro Nobili-Vitelleschi, Messrs. A. J . Cotton and R. Atkinson, B. King and others. They have located substantial evidence of mesolithic and Late Palaeolithic activity over the whole area. There are also later prehistoric elements in all these areas, but I intend, in this paper, to give only figures for artifacts which are of a mesolithic aspect. Mention will also be made of a few Late Palaeolithic specimens. Surface collections from Hengistbury Head are prodigious but the artifacts listed can be regarded as representative.

Warren Hill This is the highest part of Hengistbury Head and is centred round SZ 178905. The stratigraphy here is: peaty sand below the turf, then leached aeolian sand with colloidal peat, concreted sand with ironpan and at the base thick deposits of ferruginous Bracklesham sand.

18


South Fields These fields are centred round SZ 160912 and are south of the Broadway, bordering on the cliff" edge and just outside the Iron Age promontory fort. The easternmost of these fields, nearest Warren Hill, is also known locally as the Long Field. The topsoil varies from sandy to clay with flints.

North Fields The fields centred round SZ 164913 are here referred to as the North Fields. They are noith of the Broadway and face Mother Siller's Channel across the harbour. They include the Wick area where Mr. Cotton has located a site with strong Late Palaeolithic character. These fields also contain a Late Neolithic element with many petit tranchet derivative arrowheads.

Harbour Foreshore Artifacts are found along the whole length of the shore below Warren Hill. The finds from this area are predominantly mesolithic so far with the exception of two Late Palaeolithic tanged points and blades, a Middle Palaeolithic scraper and flake and a small Levallois core. Some of these artifacts may have been derived from the top of Warren Hill, but some have been found sticking out of the shallow muddy bank. Surface collecting in this area has not been as frequent as on the fields and Warren Hill. This explains the smaller figures for artifacts from here. No microliths have yet been found but this may be due to the muddy conditions.

HENGISTBURY HEAD ARTIFACTS

WARREN

HILL

NORTH

FIELDS

SOUTH

FIELDS FORESHORE TOTAL

Microliths: Obliquely blunted (Ai) Obliquely blunted with opposite retouch (Aa) Blunted whole of one

7

1

4 2 r 3

1

edge (Bi) Blunted whole of two

11 3 2 16

edges (B2) Blunted one edge and

1 2 3

across base (Ci) Sub-triangular (D5) Point with inverse

3 1 2

4 2

retouch on the base (E) Microlithic tanged point (G) Transverse arrowhead

3

1

3

1

(H) Unclassified (Broken)

Axes, core 2

2

1 2 1 1

1

4 4

Flake axes 1 1

Limpet-hammers 2 2

19

PROCEEDINGS FOR THE YEAR IQ70

W A R R E N N O R T H SOUTH

H I L L FIELDS FIELDS FORESHORE T O T A L

Scrapers: Side 3 1 1 2 6 l5 1 7 2

Two sides 3 1 4 3 2 0

End 7 0 79 9 2 1 6 0

Side/end 9 5 1 4

Double end 1 9 5 15 Steep on chunky flakes 1 2 4 1 3 0 3 2 11 2 9 7 Convex 3 0 2 448 2 1 1 ' 7 578 Core 12 12

Micro-scrapers 16 2 0 8 2 46 Concave 3 2 3 8 34

Tanged points 7 4 1 2 1 4 Shouldered points 4 6 1 11

Truncated flakes and blades 1 4 1 6 Blunted flakes 12 2 5 7 2 46 Blunted blades J 3 5 1 1 2 0

Gravers 2 4 10 2 36 Knives 8 34 4 46 Borers or awls 6 21 4 3i Saws 2 2 5 9 Flakes: retouched and utilised

(estimated) 1,500 5 0 0 5 0 0 2 0 0 2 ,700

Blades: retouched and utilised (estimated) 6 0 0 2 0 0 2 0 0 8 0 1,080

Micro-burins: bulbar and tip 1 4 5 Krukowski 1 1

Small blade-cores 33 53 16 3 0 1 3 2

Core-trimming flakes (estimated) 4 0 0 3 0 0 3 3 0 5 0 1,080 Tranchet axe trimming flakes 1 1 2 Waste flakes and blades

estimated 5,000 5,000 3 ,000 5 0 0 13.500

Microlitlis and Other Blunted Forms I t may be significant that on Warren Hill the most common microlith is form Bi which is blunted down the whole of one side (16 out of 48). Out of 105 microlithic blunted blades from the Upper Palaeolithic site of La Madeleine (British Museum coll.) 64 are blunted down one side. I t is therefore possible that some of the microliths of this form from Warren Hill may in fact belong to the Late Palaeolithic industry on the Head. This suspicion is strengthened by the fact that six of the blunted blades are of microlithic dimensions although they still retain their bulbs of percussion and are therefore technically not microliths made by the notch technique. This also applies to three from the North Fields (fig. 12:2, 8; fig. 14:22). The Late Palaeolithic industry on the Head is often patinated a characteristic blue and this is found in one or two of the microliths of Form Bi .

Tanged and Shouldered Points It is probable that most, if not all, the tanged and

20

Fig. 11. Flint implements from Mother Siller's Channel.

Fig. 12. Flint implements from the Bournemouth and Christchurch area.

0 '



shouldered points are Late Palaeolithic. A study of these artifacts suggests that they can be divided into two general groups. The first group has a characteristic blue patination and the points are mostly made on elegant blades; in the second group the points are frequently fairly thick, frequently have inverse retouch and have a patination no different from that of the mesolithic artifacts. This could possibly indicate that more than one Late Palaeolithic industry existed on the Head. Without stratigraphical evidence the chronological relationship between the latest Palaeolithic and the earliest mesolithic of the area cannot be established.

Axes and Axe Trimming Flakes One of the axes from Warren Hill in the Red House Museum is of the non-tranchet type, the specimen from the South Field (marked Long Field) is of tranchet type and the specimen from the North Field cannot be classified as it consists of the butt end only. I t has not been possible to see the second axe from Warren Hill, but I have been informed that it is of the tranchet type.

Scrapers Convex scrapers are common in mesolithic industries but also in later industries. Mesolithic forms are generally light and with small neat retouch. They often retain some cortex. Steeply retouched forms on thick or irregular, chunky flakes are characteristic.

Micro-burins Even though the statistical analysis is mainly based on surface collecting, the microlith/micro-burin ratio of nearly 8:1 is surprising. Most mesolithic assemblages usually have more micro-burins than microliths. This again raises the question whether all the microliths were made by the typical mesolithic notch technique.

Krukowski micro-burins are the resulting waste pieces when a finished microlith is utilised as a primary blade for the production of another microlith. The artifact would therefore have one or more blunted edges as well as a micro-burin facet (de Sonneville-Bordes i960, n - 1 2 ) . These artifacts occur in Upper Palaeolithic contexts in France and in England, as for instance from Charterhouse-on-Mendip, a Creswellian site (Pitt-Rivers Mus., Oxford). They also occur fairly frequently in mesolithic assemblages but have not always been recognised, e.g. one from Hastings (British Mus.); six were found on Portland Site 1 (Palmer 1969) and quite a few have so far been found at the nearby Culver Well site. Three were found at Three Holes Cave, Torbryan (Rosen-feld 1964, 15).

MINOR MESOLITHIC SITES : BOURNEMOUTH AREA

Most of the artifacts from the area appear to be from surface collections. Many of these collections were made at the beginning of the century when large areas were still available for examination. The finds can therefore be regarded as reasonably representative of prehistoric activities in the area. It is not always possible to locate find-spots exactly and Grid References will be followed by an indication whether they are accurate (A), general (G) or merely estimated (E).

Avon Common. SZ 130986 (G). Three micro-burins, 1 graver and 3 or 4 cores from this area are mesolithic. The common is near the River Avon.

24


Parley Cross. SZ 083983 (G). This site is on the Dorset border. The diagnostic artifacts are: 1 heavy tranchet axe about 12 in. long; 1 convex scraper and 1 steep scraper on a chunky flake.

Pokesdown. SZ 125922 (G). Two retouched blades, one is 13.2 cm. long. They are marked 'Plateau above gravel' and may be Late Palaeolithic. They are patinated a rich brown. More typically mesolithic are 4 convex scrapers and a retouched blade segment. There are two microliths from this site in the Red House Museum, one with one side and the base blunted (form Ci) and the other an elongated trapeze. The trapeze is rather large for a microlith and approaches a Cheddarian point (fig. 11, no. 7).

Queen's Park. SZ 110935 (G). One microlith, an isosceles triangle, all three sides blunted, the right side being inversely retouched; 1 tranchet axe.

Red Hill Common. SZ 088956 (G). This is a mixed site with a Bronze Age element. Diagnostically mesolithic are: 1 micro-core with one platform partly utilised; 2 retouched flakes which clearly came from micro-cores, and a few possible waste flakes and blades.

Sopley Common. SZ 130977 (G). This is yet another site near the River Avon. Mesolithic material includes: 3 micro-burins; 2 convex scrapers; 3 retouched blades; 5 micro-cores; 43 waste flakes and 3 waste blades. A lot of the artifacts are burnt. This could indicate the presence of a habitation site.

Southbourne (26 Clifton Road). SZ 136914 (A). One densely patinated flake with later blunting retouch in the mesolithic style. The blunting occurs at the bulbar end of the flake. The original flake may very well have been Palaeolithic. A small notched blade and a retouched flake from the same garden also have a mesolithic appearance.

Talbot Wood. SZ 070940 (G). Most of this area is now built over with the exception of a few small patches of trees. There is however every indication that a site of fair importance once existed in this general area. Unfortunately some of the finds in collections are merely marked 'Talbot Wood' but it seems as though a good proportion came from Glenfernes Avenue and a few from Huntly Road, parallel to Glenfernes (sometimes spelt Glenfurness). There are elements of Early and Late Palaeolithic, Mesolithic and Bronze Age aspects in the assemblages, but they may not all have come from the same place. Known to have come from Glenfernes Avenue (from 'loam below soil') are: 1 utilised blade 12 cm. long; 1 retouched blade and 1 flake from a micro-core; 1 saw; 1 knife on a blade 9 cm. long. From the general area of Talbot Wood the following artifacts are of a mesolithic aspect: 2 cores; 5 core trimming flakes; 1 graver; 5 convex scrapers; 1 scraper on a chunky flake; 1 Portland chert utilised flake; 6 other utilised flakes; 1 utilised blade; 1 borer and possibly 1 knife.

Tackton. SZ 145923 (G). Three tranchet axes, 2 core trimming flakes and 3 long blade-cores are in the Red House Museum. It is not known whether all these artifacts were found together. Tuckton is near the harbour.

Wallis Down. SZ 070945 (G). This area adjoins Talbot Wood and may in fact be the same site, or an extension of it. One saw and 1 retouched blade from here have been noted.

West Way (lower end). SZ 099944 (E). One microlith: obliquely blunted along part of one side (form Ai ) .

25

PROCEEDINGS FOR THE YEAR I 9 7 O

STRAY FINDS OF AXES AND PICKS FROM BOURNEMOUTH

AXES, AXES, PICKS

ADZE TRANCHET NON-TRANCHET PICKS FORM

Barrow Plot, Wick Lane. SZ 154922 (E) 3 1 1

Boscombe. SZ 120915 (E) 1

Boscombe. King's Park. SZ 116925(G) 1 1

East Cemetery. SZ 120926 (A) 1

Hum. Fillybrook Farm. SZ 130970 (E) 1

Southbourne. SZ 140912 (G) 1 1

Winton. SZ 085935 (G) 1

Locality unknown 1

8 2 3 1

The Barrow Plot site is near the North Fields, Hengistbury Head. The large implement from Boscombe (now in the British Museum) was found six feet down in gravel between Boscombe and Pokesdown. It is patinated a deep rust-brown. It has a flat under-surface and a broad working edge and would be more suitable for use as an adze rather than as an axe or pick. The axe from Southbourne has a finely retouched working edge, convex-concave in section and is one of the type sometimes described as a 'Nostvet' axe (Troels-Smith 1937, 281).

CHRISTCHURCH AREA Furzy. SZ 155939 (G). Most, if not all, the artifacts from this area are from Latch

Farm. It is obviously an important site, near the banks of the River Avon. Later prehistoric material has also been found in the same area (Calkin 1966). Mesolithic artifacts are:

Microliths: obliquely blunted (Ai) Scrapers: side

end-on-flake/blade side-and-end

2

55 8 8

steep-on-chunky flakes 44 convex 440 micro-scrapers 8 concave 1

Tranchet axes 4 Picks (non-tranchet) Shouldered points Truncated blades

2 1 2

Blunted flakes 1

26


Knives 4 Utilised/retouched flakes 148 Utilised/retouched blades 39 Micro-cores 33 Waste flakes/blades (mesolithic aspect) 150

One of the end-scrapers is on a blade 7.7 cm. long, is steeply retouched (about 90°) with undercutting of the working edge. It could be Late Palaeolithic. This also applies to the shouldered point.

Mill Plain. SZ 158935 (E). This site is also near the River Avon. Four microliths, a micro-burin and 2 tranchet axes are known from here. The microliths are: 1 obliquely blunted (form Ai), 1 blunted down the whole of one edge with a small area of retouch opposite (Bi), a scalene triangle (Dib), and one with inverse retouch on the base (formE).

Bosley. SZ 147958 (E). One convex end-on-blade scraper may be Late Palaeolithic. It is 11 cm. long. Some mesolithic industries do have very long blades, as for instance at Hastings (British Mus.) or Cliff End, Sussex (Palmer, forthcoming).

Homelands, Douglas Avenue. SZ 150926 (A). This site is near Tuckton Bridge, on the River Avon, not far from where the river flows into the Christchurch Harbour. Artifacts from here: 1 tranchet axe trimming flake; 3 micro-burins (2 bulbar, 1 tip variety); 2 core-trimming flakes; 1 utilised flake and 43 waste flakes. This site is important as it is near to both Hengistbury Head and Mother Siller's Channel.

King's Avenue. SZ 160928 (E). A long, very unusual lanceolate point in the Red House Museum is difficult to date (fig. 12, no. 9). It is 6.5 cm. long and is blunted along both edges. The blunting is in the microlithic style, but the tool could equally well be Late Palaeolithic. The find-spot is also near Tuckton Bridge, near the Homelands site.

Winkton Mead. SZ 168958 (G). One utilised blade, 6.7 cm. long. Highcliff. Rankine (1956) mentions microliths. No details are known.

STRAY FINDS OF AXES AND PICKS FROM CHRISTCHURCH

AXES,

TRANCHET

AXES,

NON-TRANCHET PICKS

Fairfield. SZ 156933 (G) Grove Farm. Unlocated. Not far from Latch Farm Harbour. No locality. Centred. SZ 170910 Mudeford. Sandhills. SZ 189921 (G) Hinton Admiral. SZ 203950 (G) Hardwell Cliffs. Unlocated Walkford. SZ 220946 (G) Locality unknown

1

1

1

1

1

7

1

1

1

1

6

12 2 8

27


T H E M E S O L I T H I G I N D U S T R I E S

ANALYSES OF ARTIFACT STATISTICS

MICROLITHS

AI A2 BI B2 CI D I B °5 E F O H UNCLASSIFIED

Mother Siller's Channel

Hengistbury Head Other sites

i

3

I

1

16 I

I

3

I

4 1

I

I 2 3

1

I I 4

i7 I 18 4 6 2 2 4 I I 4

The obliquely blunted form Ai is common on mesolithic sites, but a high figure for form Bi is unusual. Most of these microliths came from Hengistbury Head and they have already been commented on. I have found that microlithic transverse arrowheads (form H, trapezes) are seldom found in mesolithic assemblages from the southern littoral areas. One of the examples from Portland Site 1 may in fact be similar to a Cheddarian point as there is evidence of a Late Palaeolithic element on that site (Palmer 1969, fig. 8, no. 46). Transverse arrowheads continued in use during the Neolithic period in the modified non-microlithic petit tranchet derivative form. Many specimens of the derivative forms have been found in the Hengistbury Head area. Sixty microliths from the whole Bournemouth and Christchurch area is a small figure, but there may be a few more in other private collections, particularly from the Head. Microliths are not easy to find during surface collecting.

Axes and Picks

TRANCHET

AXES

NON-TRANCHET

AXES PICKS

FLAKE

AXES ADZES

UNCLAS

SIFIED

A X E

TRIMMING FLAKE

Mother Siller's Channel

Hengistbury Head

Other sites 2

3 1

1

4 ! 3

1

1

1

2

2 1

33 5 ' 3 1 1 1 5

Total Axes and Picks: 54

The figures for axes and picks indicate an eco-system in which cutting down shrubs and digging up edible plants and roots were important activities. In the coastal areas of the Hampshire Basin and adjoining counties the distribution of axes and pointed picks, particularly, appears to be more related to areas which are at present fairly

29


barren except for bushes. It is not always possible to determine whether these areas were also barren in prehistoric times as various factors can affect tree-growth, but the axe and pick distribution does raise the query whether these tools were primarily intended for adaptation to forest conditions.

Gravers The 38 gravers listed are probably all mesolithic, even though 36 were found on

Hengistbury Head. Most of the Palaeolithic gravers from the Head are densely patinated and are on long blades.

FOLK-MOVEMENT INTO THE BOURNEMOUTH-CHRISTCHURCH AREA

The best evidence of folk-movement into the area during the mesolithic period is afforded by the presence of Portland chert artifacts in the assemblages listed:

Mother Siller's Channel: Total 4. See excavation report, above, for details. Hengistbury Head: 6 retouched flakes; 1 tanged point with both sides of the tang

blunted, on a flake with pronounced hinge fracture (Late Palaeolithic ?); 1 blade with blunting retouch; 2 core-trimming flakes; 1 retouched blade. Total: 11.

Talbot Wood: 1 utilised flake. Furzy, Latch Farm: 2 convex scrapers; 2 utilised flakes; 1 micro-core, multiplatform;

1 waste flake. Total: 6. This makes a total of 20 Portland chert artifacts in the area. This evidence suggests

connections with the prolific mesolithic industries on the Isle of Portland. Small bands of folk from Dorset apparently wandered far and wide along the coast and inland along the rivers.

SUMMARY AND CONCLUSIONS

1. A scatter of mesolithic artifacts have been found distributed over most of the Bournemouth-Christchurch area.

2. Most of the larger sites appear to be near the harbour or by a river. There are still large areas available for search, bordering the built-up areas. I t is likely that most of the mesolithic folk in the area followed a coastal economy.

3. On several of the mesolithic sites artifacts have been found with a Late Palaeolithic character. Several of the mesolithic tool-types and cores are almost indistinguishable from Palaeolithic specimens. This raises the question whether there may have been any relationship between the two industries. Stratigraphical evidence is still lacking.

4. There was contact between the mesolithic folk of the Bournemouth-Christchurch area and the mesolithic coastal dwellers of Dorset.

Acknowledgements I am indebted to the Christchurch Urban District Council and the Stanpit Marsh

Local Nature Reserve Management Committee for permission to conduct the excavation at Mother Siller's Channel.

I further wish to acknowledge the co-operation of the following people: the Curator of the Red House Museum, Christchurch; the Keeper and staff of the British Museum

30

THE MESOLITHIG INDUSTRIES

(Department of Prehistoric and Romano-British Antiquities); Mr. A. M. ApSimon (Southampton University), Mr. R. Atkinson, Mr. and Mrs. A. J . Cotton, Mr. Bruce King, Marchese Alessandro Nobili-Vitelleschi and the excavation volunteers. I am grateful to Dr. I. W. Cornwall of the Institute of Archaeology for his comments on the soil samples.

APPENDIX A

ANALYSIS OF SAMPLES FROM SECTION OF

MOTHER SILLER'S CHANNEL, CHRISTCHURCH HARBOUR

By I. W. CORNWALL

At about 3 ft. above present-day H W M , or reached only by the very highest tides, the site must have been some quite considerable height above the sea in Mesolithic times, if one may judge by the general low level of the sea assigned to comparable sites in the North Sea basin.

i. The sand was exclusively quartz and the individual grains in all the samples were of the same character — only subrounded for the most part and highly polished. There were only very few well-rounded grains in any sample and these, too, had polished surfaces. They may, once, have been wind-borne and matt-surfaced, but have, latterly, obviously been water-transported for a long time. Not a single typically wind-worn grain was seen.

In the Christchurch region, the source of most sands, not of recent marine origin, would be the local Eocene deposits of the Hampshire Basin, probably Reading or Bagshot Beds, the intervening London Clay having but a very small sand-content. Both are marine, and the sand-grains would have acquired their characteristic water-polish on Eocene sea-beaches. The absence of any wind-worn grains at all suggests that all the material is of pre-glacial geological origin.

2. Of more interest, perhaps, is the fact that Layer 3 contains many still-recognisable plant-remains and more humus than the rest, and so looks to be an ancient surface bearing a podsol-type soil-profile. This has been buried subsequently by the sands of Layers 1 and 2. On these, in turn, a less distinct and mature podsol-profile has developed since the burial. The presence of this soil-profile with its implication of probably close vegetation-cover, practically rules out any but the most occasional wetting with salt water, which would have killed the vegetation at once. Unless there is any adjacent slope from which sand might have been deposited by rain-washing, one is left with wind as the most likely agency by which the ancient surface became buried, though this cannot be discerned in the samples. As the sand-grains are of identical character throughout the section, it looks as if the materials were entirely local, and of the same origin, and, whether transported by water or wind, have not travelled far. Such brief wind-transport would have no perceptible effect on the sand-grains, in which the acquisition of the well-rounded form and matt surface typical of dune-sands takes at least some thousands, if not tens or hundreds of thousands, of years to develop - and as many more to erase, if the grains again fall into water.

31


I would reconstruct the section from the samples as follows:

| | | | I I I I I I A l MODERN S U R F A C E

2 A>

• l l l l l l l l l l l l l l l l l l l A ^ ^ V . r.L«A»M,T» S U R F A C E )

* A2 5 %%%%%%%%%%%&$)£- P E B B L Y F E R R U G I N O U S

Kg- '5

The upper half of (3) contains more iron than the layer above it and its own lower part and (4). This might be interpreted as the B-horizon of the upper soil-profile. Minerals are often deposited at an old land-surface, perhaps because of its denser texture.

The profile of the old soil has no B-horizon distinguishable from these samples, but the iron may well have percolated for some distance into the subsoil before being precipitated, as is not uncommon on such pervious sediments. I would expect to find an iron-pan at, or about, the level of the water-table.

REFERENCES Calkin, J. B. (1966). Discovering Prehistoric Bournemouth and Christchurch. Christchurch, Red

House Museum. De Sonneville-Bordes, D. (i960). La Paleolithique Superieure en Perigord. Bordeaux. Godwin, H., and Willis, E. H. (1959). 'Cambridge University Natural Radiocarbon Measure

ments 1', Radiocarbon, 1, 63-75. Mace, Angela (1959). 'An Upper Palaeolithic Open Site at Hengistbury Head', Proc. Prehist.

Soc, 25, 233-259. Palmer, Susann (1969). 'A Mesolithic Site at Pordand Bill, 1966', Proc. Dorset Nat. Hist, and

Arch. Soc, go, 183-206 and 163. Rankine, W. F. (1956). The Mesolithic of Southern England, Surrey Arch. Soc. Paper, No. 4. Rosenfeld, A. (1964). 'Excavations in the Torbryan Caves, Devonshire. II. Three Holes

Caves', Trans. Devon Arch. Expl. Soc, 22, 3-26. Troels-Smith, J. (1937). 'Beile aus dem Mesolithicum Danemarks. Ein Einteilungsversuch',

Acta Archaeologica, 8, 278-291.

32

Documents

THE MESOLITHIG INDUSTRIES OF MOTHER SILLER'S CHANNEL ... · 1. The Mother Siller's site appears to have been a camping site during the meso-lithic period, possibly afte B.Cr whe 500